Slurm scheduler in AWS PCS

Slurm is an open-source workload manager designed for Linux clusters that provides job scheduling, resource allocation, and job monitoring capabilities for HPC workloads. AWS PCS supports the Slurm scheduler to manage your cluster workloads.

Topics

• Slurm versions in AWS PCS

• Slurm configuration in AWS PCS

• Slurm accounting in AWS PCS

• Slurm REST API in AWS PCS

• Rebooting compute nodes with Slurm in AWS PCS

• Configuring hardware topology in AWS PCS

• Configuring Multi-Instance GPU (MIG) in AWS PCS

• Extend Slurm functionality on AWS PCS with SPANK plugins

• Use Slurm CLI Filter Plugins to customize job submission in AWS PCS

• Slurm metrics in AWS PCS

• Collect Slurm metrics with a managed Prometheus collector